mathdatech/twoot
1
0
Fork 0
mirror of https://gitlab.com/jeancf/twoot.git synced 2024-11-30 15:31:11 +00:00
Code Issues 1 Packages Projects Releases Wiki Activity

Merge branch 'video-dl'

Browse Source
This commit is contained in:
JC Francois 2020-04-05 10:39:43 +02:00
commit f635fb5eda
4 changed files with 152 additions and 67 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
.gitignore vendored
View File

@ -2,4 +2,5 @@
.vscode/
venv/
*.secret
*.sh
twoot.db

34
README.md
View File

@ -1,9 +1,17 @@
# Twoot
Twoot is a python script that extracts tweets from a twitter feed and
reposts them as toots on a Mastodon account.
I started twoot when [tootbot](https://github.com/cquest/tootbot)
stopped working. Tootbot relies on rss feeds from https://twitrss.me
that broke when Twitter refreshed their web UI in July 2019.
Instead twoot is self contained and handles all the processing.
**UPDATE 05 APR 2020** VERSION 1.0. Twoot can now optionally download
videos from Twitter and upload them on Mastodon.
**UPDATE 17 MAR 2020** Added command line switch (`-r`) to also post
reply-to tweets on the mastodon account. They will not be included by
default anymore.
@ -20,17 +28,18 @@ of last week.
* Fetch timeline of given users from twitter.com
* Scrape html and formats tweets for post on mastodon
* Emojis supported
* Upload images from tweet to Mastodon (videos not supported)
* Optionally upload videos from tweet to Mastodon
* Upload images from tweet to Mastodon
* Specify maximum age of tweet to be considered
* Specify minimum delay before considering a tweet for upload
* Remember tweets already tooted to prevent double posting
* Can optionally post reply-to tweets on the mastodon account
* Optionally post reply-to tweets on the mastodon account
# usage
```
twoot.py [-h] -t <twitter account> -i <mastodon instance> -m <mastodon
account> -p <mastodon password> [-r]
account> -p <mastodon password> [-r] [-v]
[-a <max age in days>] [-d <min delay in mins>]
```
@ -41,14 +50,20 @@ is @superduperbot@botsin.space
|Switch |Description | Example | Req |
|-------|--------------------------------------------------|--------------------|-----|
| -t | twitter account name without '@' | `SuperDuperBot` | Yes |
| -t | twitter account name without '@' | `SuperDuper` | Yes |
| -i | Mastodon instance domain name | `botsin.space` | Yes |
| -m | Mastodon username | `superduperbot` | Yes |
| -p | Mastodon password | `my_Sup3r-S4f3*pw` | Yes |
| -v | upload videos to Mastodon | *N/A* | No |
| -r | Post reply-to tweets (ignored by default) | *N/A* | No |
| -a | Max. age of tweet to post (in days) | `1` | No |
| -a | Max. age of tweet to post (in days) | `5` | No |
| -d | Min. delay before posting new tweet (in minutes) | `15` | No |
When using the `-v` switch consider:
* whether the copyright of the content that you want to cross-post allows it
* the storage / transfer limitations of the Mastodon instance that you are posting to
* the upstream bandwidth that you may consume on your internet connection
Default max age is 1 day. Decimal values are OK.
Default min delay is 0 minutes.
@ -57,8 +72,15 @@ Default min delay is 0 minutes.
Make sure python3 is installed.
Twoot depends on beautifulsoup4 and mastodon python module: `sudo pip install beautifulsoup4 Mastodon.py`
Twoot depends on `beautifulsoup4` and `Mastodon.py` python modules.
**Only If you plan to download videos** with the `-v` switch, are the additional dependencies required:
* Python modules `m3u8` and `ffmpeg-python`
* [ffmpeg](https://ffmpeg.org/download.html) (installed with the package manager of your distribution)
```
> pip install beautifulsoup4 Mastodon.py m3u8 ffmpeg-python
```
In your user folder, execute `git clone https://gitlab.com/jeancf/twoot.git`
to clone repo with twoot.py script.

24
twitterdl.py Normal file → Executable file
View File

@ -1,6 +1,28 @@
#! /usr/bin/env python3
"""
This file is a modification of
https://github.com/h4ckninja/twitter-video-downloader/
The original package has an unknown license. The modified version
is released here under GPL v3.
Copyright (C) 2019 Jean-Christophe Francois
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program. If not, see <http://www.gnu.org/licenses/>.
"""
import argparse
import requests
import json
import urllib.parse

132
twoot.py
View File

@ -1,7 +1,7 @@
#! /usr/bin/env python3
# -*- coding: utf-8 -*-
'''
"""
Copyright (C) 2019 Jean-Christophe Francois
This program is free software: you can redistribute it and/or modify
@ -16,7 +16,7 @@
You should have received a copy of the GNU General Public License
along with this program. If not, see <http://www.gnu.org/licenses/>.
'''
"""
import sys
import argparse
@ -29,8 +29,9 @@ import datetime, time
import re
from pathlib import Path
from mastodon import Mastodon, MastodonError, MastodonAPIError, MastodonIllegalArgumentError
import twitterdl
import subprocess
import json.decoder
import shutil
# Update from https://www.whatismybrowser.com/guides/the-latest-user-agent/
@ -84,11 +85,17 @@ def handle_no_js(session, page, headers):
return new_page
def cleanup_tweet_text(tt_iter):
def cleanup_tweet_text(tt_iter, twit_account, status_id, tweet_uri, get_vids):
'''
Receives an iterator over all the elements contained in the tweet-text container.
Processes them to remove Twitter-specific stuff and make them suitable for
posting on Mastodon
:param tt_iter: iterator over the HTML elements in the text of the tweet
:param twit_account: Used to name directory where videos are downloaded
:param status_id: Used to name directory where videos are downloaded
:param tweet_uri: Used to downloaded videos
:param get_vids: True to download embedded twitter videos and save them on the filesystem
'''
tweet_text = ''
# Iterate elements
@ -120,14 +127,22 @@ def cleanup_tweet_text(tt_iter):
if tag.has_attr('data-expanded-path'):
data_expanded_path = tag['data-expanded-path']
if 'video' in data_expanded_path:
# Download video from twitter and store in filesystem
tweet_uri = "https://twitter.com/" + data_expanded_path.strip("/video/1")
twitter_dl = twitterdl.TwitterDownloader(tweet_uri, target_width=500, debug=1)
if get_vids:
# Download video from twitter and store in filesystem. Running as subprocess to avoid
# requirement to install ffmpeg and ffmpeg-python for those that do not want to post videos
try:
twitter_dl.download()
except json.JSONDecodeError:
print("ERROR: Could not get playlist")
# TODO set output location to ./output/twit_account/status_id
dl_feedback = subprocess.run(
["./twitterdl.py", tweet_uri, "-ooutput/" + twit_account + "/" + status_id, "-w 500"],
capture_output=True
)
if dl_feedback.returncode != 0:
# TODO Log dl_feedback.stderr
tweet_text += '\n\n[Video embedded in original tweet]'
except OSError:
print("Could not execute twitterdl.py (is it there? Is it set as executable?)")
sys.exit(-1)
else:
tweet_text += '\n\n[Video embedded in original tweet]'
# If element is hashflag (hashtag + icon), handle as simple hashtag
@ -177,15 +192,15 @@ def contains_class(body_classes, some_class):
def main(argv):
# Build parser for command line arguments
# TODO Add option for ingestion of video content
parser = argparse.ArgumentParser(description='toot tweets.')
parser.add_argument('-t', metavar='<twitter account>', action='store', required=True)
parser.add_argument('-i', metavar='<mastodon instance>', action='store', required=True)
parser.add_argument('-m', metavar='<mastodon account>', action='store', required=True)
parser.add_argument('-p', metavar='<mastodon password>', action='store', required=True)
parser.add_argument('-r', action='store_true')
parser.add_argument('-a', metavar='<max age in days>', action='store', type=float, default=1)
parser.add_argument('-d', metavar='<min delay in mins>', action='store', type=float, default=0)
parser.add_argument('-r', action='store_true', help='Also post replies to other tweets')
parser.add_argument('-v', action='store_true', help='Ingest twitter videos and upload to Mastodon instance')
parser.add_argument('-a', metavar='<max age (in days)>', action='store', type=float, default=1)
parser.add_argument('-d', metavar='<min delay (in mins)>', action='store', type=float, default=0)
# Parse command line
args = vars(parser.parse_args())
@ -195,9 +210,16 @@ def main(argv):
mast_account = args['m']
mast_password = args['p']
tweets_and_replies = args['r']
get_vids = args['v']
max_age = float(args['a'])
min_delay = float(args['d'])
# Try to open database. If it does not exist, create it
sql = sqlite3.connect('twoot.db')
db = sql.cursor()
db.execute('''CREATE TABLE IF NOT EXISTS toots (twitter_account TEXT, mastodon_instance TEXT,
mastodon_account TEXT, tweet_id TEXT, toot_id TEXT)''')
# **********************************************************
# Load twitter page of user. Process all tweets and generate
# list of dictionaries ready to be posted on Mastodon
@ -239,14 +261,30 @@ def main(argv):
# Verify that we now have the correct twitter page
body_classes = soup.body.get_attribute_list('class')
assert contains_class(body_classes, 'users-show-page'), \
'This is not the correct twitter page. Quitting'
assert contains_class(body_classes, 'users-show-page'), 'This is not the correct twitter page. Quitting'
# Replace twit_account with version with correct capitalization
twit_account = soup.find('span', class_='screen-name').get_text()
# Extract twitter timeline
timeline = soup.find_all('table', class_='tweet')
for status in timeline:
# Extract tweet ID and status ID
tweet_id = str(status['href']).strip('?p=v')
status_id = tweet_id.split('/')[3]
# Check in database if tweet has already been posted
db.execute('''SELECT * FROM toots WHERE twitter_account = ? AND mastodon_instance = ? AND
mastodon_account = ? AND tweet_id = ?''',
(twit_account, mast_instance, mast_account, tweet_id))
tweet_in_db = db.fetchone()
if tweet_in_db is not None:
# Skip to next tweet
continue
reply_to_username = None
# Check if the tweet is a reply-to
reply_to_div = status.find('div', class_='tweet-reply-context username')
@ -259,9 +297,6 @@ def main(argv):
# Skip this tweet
continue
# Extract tweet id
tweet_id = str(status['href']).strip('?p=v')
# Extract url of full status page
full_status_url = 'https://mobile.twitter.com' + tweet_id + '?p=v'
@ -305,7 +340,7 @@ def main(argv):
authenticity_token = soup.find('input', {'name': 'authenticity_token'}).get('value')
form_input = {'show_media': 1, 'authenticity_token': authenticity_token, 'commit': 'Display media'}
full_status_page = session.post(full_status_url + '?p=v', data=form_input, headers=headers)
full_status_page = session.post(full_status_url, data=form_input, headers=headers)
# Verify that download worked
assert full_status_page.status_code == 200, \
@ -338,7 +373,7 @@ def main(argv):
# extract iterator over tweet text contents
tt_iter = tmt.find('div', class_='tweet-text').div.children
tweet_text = cleanup_tweet_text(tt_iter)
tweet_text = cleanup_tweet_text(tt_iter, twit_account, status_id, full_status_url, get_vids)
# Mention if the tweet is a reply-to
if reply_to_username is not None:
@ -381,12 +416,18 @@ def main(argv):
pass
# Check if video was downloaded
sid = re.search('/([0-9]+)$', tweet_id)
status_id = sid.groups()[0]
video_path = Path('./output') / author_account / status_id
video_file_list = list(video_path.glob('*.mp4'))
video_file = None
video_path = Path('./output') / twit_account / status_id
if video_path.exists():
# Take the first subdirectory of video path (named after original poster of video)
video_path = [p for p in video_path.iterdir() if p.is_dir()][0]
# Take again the first subdirectory of video path (named after status id of original post where vidoe is attached)
video_path = [p for p in video_path.iterdir() if p.is_dir()][0]
# list video files
video_file_list = list(video_path.glob('*.mp4'))
if len(video_file_list) != 0:
# Extract posix path of first video file in list
video_file = video_file_list[0].absolute().as_posix()
# Add dictionary with content of tweet to list
@ -407,16 +448,10 @@ def main(argv):
# print(t)
# **********************************************************
# Iterate tweets. Check if the tweet has already been posted
# on Mastodon. If not, post it and add it to database
# Iterate tweets in list.
# post each on Mastodon and reference to it in database
# **********************************************************
# Try to open database. If it does not exist, create it
sql = sqlite3.connect('twoot.db')
db = sql.cursor()
db.execute('''CREATE TABLE IF NOT EXISTS toots (twitter_account TEXT, mastodon_instance TEXT,
mastodon_account TEXT, tweet_id TEXT, toot_id TEXT)''')
# Create Mastodon application if it does not exist yet
if not os.path.isfile(mast_instance + '.secret'):
try:
@ -450,17 +485,6 @@ def main(argv):
# Upload tweets
for tweet in reversed(tweets):
# Check in database if tweet has already been posted
# FIXME Move tests to the front of the process to avoid the unnecessary processing of already ingested tweets
db.execute('''SELECT * FROM toots WHERE twitter_account = ? AND mastodon_instance = ? AND
mastodon_account = ? AND tweet_id = ?''',
(twit_account, mast_instance, mast_account, tweet['tweet_id']))
tweet_in_db = db.fetchone()
if tweet_in_db is not None:
# Skip to next tweet
continue
# Check that the tweet is not too young (might be deleted) or too old
age_in_hours = (time.time() - float(tweet['timestamp'])) / 3600.0
min_delay_in_hours = min_delay / 60.0
@ -470,8 +494,18 @@ def main(argv):
# Skip to next tweet
continue
# Upload photos
media_ids = []
# Upload video if there is one
if tweet['video'] is not None:
try:
media_posted = mastodon.media_post(tweet['video'])
media_ids.append(media_posted['id'])
except (MastodonAPIError, MastodonIllegalArgumentError, TypeError): # Media cannot be uploaded (invalid format, dead link, etc.)
pass
else: # Only upload pic if no video was uploaded
# Upload photos
for photo in tweet['photos']:
media = False
# Download picture
@ -511,6 +545,12 @@ def main(argv):
(twit_account, mast_instance, mast_account, tweet['tweet_id'], toot['id']))
sql.commit()
# Cleanup downloaded video files
try:
shutil.rmtree('./output/' + twit_account)
except FileNotFoundError: # The directory does not exist
pass
if __name__ == "__main__":
main(sys.argv)